Upload 6 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+f-gene_verification/model/best_model.keras filter=lfs diff=lfs merge=lfs -text

f-gene_verification/Dockerfile

@@ -0,0 +1,12 @@
+FROM python:3.9
+
+WORKDIR /app
+
+COPY . /app
+
+RUN pip install --upgrade pip
+RUN pip install -r requirements.txt
+
+EXPOSE 7860
+
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

f-gene_verification/main.py

@@ -0,0 +1,73 @@
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+import pickle
+import numpy as np
+from tensorflow.keras.models import load_model
+from tensorflow.keras.preprocessing.sequence import pad_sequences
+
+# Load model artifacts
+model = load_model("model/best_model.keras")
+
+with open("model/kmer_to_index.pkl", "rb") as f:
+    kmer_to_index = pickle.load(f)
+
+with open("model/maxlen.txt", "r") as f:
+    maxlen = int(f.read().strip())
+
+# K-mer tokenizer
+def kmer_tokenizer(seq, k=6):
+    return [seq[i:i + k] for i in range(len(seq) - k + 1)]
+
+# Preprocess function
+def preprocess_sequence(sequence: str):
+    tokens = kmer_tokenizer(sequence.upper())
+    encoded = [kmer_to_index.get(kmer, 0) for kmer in tokens]
+    padded = pad_sequences([encoded], maxlen=maxlen, padding='post')
+    return padded
+
+# Define request schema
+class SequenceInput(BaseModel):
+    sequence: str
+
+# Class labels (optional)
+LABELS = ["Random", "F", "P", "N", "M", "HN", "L"]
+
+# Create app
+app = FastAPI(title="Gene Classifier API")
+
+@app.post("/predict")
+def predict(input_data: SequenceInput):
+    seq = input_data.sequence.strip()
+    if len(seq) < 6:
+        raise HTTPException(status_code=400, detail="Sequence too short. Must be at least 6 bases.")
+
+    try:
+        padded = preprocess_sequence(seq)
+        pred = model.predict(padded)
+        predicted_class = int(np.argmax(pred))
+        label = LABELS[predicted_class]
+        confidence = float(np.max(pred))
+
+        # Custom logic
+        if label == "F":
+            return {
+                "status": "success",
+                "message": "F gene detected.",
+                "confidence": confidence
+            }
+        elif label == "Random":
+            return {
+                "status": "error",
+                "message": "Unidentified sequence detected. Make sure you're entering the F gene of the NDV.",
+                "confidence": confidence
+            }
+        else:
+            return {
+                "status": "error",
+                "message": "No F gene detected. Please enter an F gene.",
+                "predicted_label": label,
+                "confidence": confidence
+            }
+
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Prediction failed: {str(e)}")

f-gene_verification/model/best_model.keras

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be7649bfcea0816dffa5cd6b2490c9c7f08bae3ec2b5e39d171ee0f661429a4e
+size 24844261

f-gene_verification/model/kmer_to_index.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd2a12751bc0417d7edd2379b2aecd79bd3fa52dafde1c376a7f604280798bcb
+size 83542

f-gene_verification/model/maxlen.txt

@@ -0,0 +1 @@
+6649

f-gene_verification/requirements.txt

@@ -0,0 +1,7 @@
+fastapi
+uvicorn
+tensorflow
+scikit-learn
+biopython
+pydantic
+numpy